I am trying to use a line of output in an XML file as input in another new XML file for processing purposes via a shell script. Since I am a newbie though, I'm not sure how to do this since the data is different everytime. I am using this technique with static data right now:

echo -n "Running data test... "
cat datafile.xml | sed -e s/changethis/intothis/ > /tmp/$$.xml
export RESULT=`~/xml /tmp/$$.xml | grep "<Code>SUCCESS</Code>"`
if [ "$RESULT" != "" ]; then { echo passed } else { echo FAILED ; exit } fi

The XML file has the data I need stored between XML tags, so I know I could grep that line and output it to another file, and use that file to refer to for the input in the next test. That file would have < Number >here'sthedata< /Number > in that file. The trick would be to parse the data out, in effect removing the tags, storing that value in a variable, and then using sed to do a replace in the new input file like I am above. I'm just unsure how to go about parsing the data out.

Hope this wasn't confusing

Well... if you don't mind going to Perl, (I would recommend this)
you can use the XML::Parser Perl module...

http://wwwx.netheaven.com/~coopercc/...ser/intro.html

...it would be alot more flexable and in the long run,
I think you'll need that anyway.

Actually a colleague here is trying to push these scripts over to Perl (mainly because he more comfortable with Perl than shell scripting). To be bluntly honest, I'm a Windows guy (ducks....) and my new job responsibilites are forcing me to pick this up on the fly. From the little exposure I've had to Perl, it probably would work more elegantly, but these testing scripts are all compelte except the one I'm having a problem on, and the ole 'time constraints' make it problematic to rewrite everything in Perl right now. I do plan on doing that eventually, but for now I need this up and running in the shell environment.

(bash bash bash)

PERL is really not so difficult to learn and there must be hundreds of samples on the net for how to build PERL::XML transformations:


PERL XML FAQ

XMLperl

And so much more:

Google search

If you REALLY, REALLY want to use awk, I guess you can
do somthing like...

...the test.xml data file is...

... the awk program will output the "data" on the
"<Number>" lines only.

I hope this encourages you to look at Perl again.

Damn...

The XML strings disappeard in the previous post...
Note I changed all the angle brackets "<" to square
brackets "[". you'll have to translate them back.

script...

#!/bin/sh

awk '/[Number]/ {stp=index($0,">")+1; \
newstr=substr($0,stp); \
etp=index(newstr,"<")-1; \
print substr(newstr,0,etp); \
}' test.xml


test.xml

[String]one[/String]
[Number]two[/Number]
[String]three[/String]
[Number]four[/Number]